Walking requires precise coordination of movement timing and scaling between the two legs. This inter-limb coordination is often disrupted after neurological injury (e.g. stroke), resulting in abnormal, asymmetric walking patterns. Recent data show that locomotor patterns can be altered through treadmill training, even after central nervous system damage, raising the possibility that abnormal inter-limb coordination could be remediated with adaptive training strategies. In this project, we will study adaptation of inter-limb coordination during walking, by using a split-belt treadmill to control walking speed of the two legs independently. Our data and the work of others suggest that inter-limb adaptation could rely, in part, on the integrity of supraspinal structures such as: 1) cerebellar-brainstem circuits and/or 2) sensorimotor regions of cerebral cortex, via the corticospinal tract. We hypothesize that inter-limb adaptation is most dependent on cerebellar-brainstem interactions. If so, people with other types of neurological damage, but with intact cerebellar function (e.g. hemiparesis), could benefit from adaptive split-belt treadmill training to correct abnormal, asymmetric walking patterns. In this project we will study locomotor adaptation mechanisms in healthy control subjects and people with focal brain damage to answer the following questions: 1) What is the human capacity for adaptation of inter-limb coordination during locomotion (dynamic range, extent of storage, and generalization to other contexts)? 2) What brain structures are critical for adaptation of inter-limb coordination during locomotion? 3) Can adaptive training improve inter-limb coordination and walking patterns in people with neurological damage? These studies will provide important new information about the neural mechanisms of locomotor adaptation, as well as providing a new rehabilitation tool for people with asymmetric gait patterns resulting from central nervous system damage (e.g. stroke, cerebral palsy). Lay summary: Walking disturbances are some of the most common problems in people who have had a stroke. In these experiments, we will work to understand how people normally coordinate the two legs to walk, how new walking patterns can be learned in short time periods, and whether people who have had damage to the brain can learn to improve walking using new rehabilitation methods. We will do this using a new type of treadmill that has two belts so that the legs can be independently controlled at different speeds.